1. Descriptive analyses

PRISMA flowchart

Location of studies

ES distribution

2. Primary analysis

Main model

# {.tabset}
dat_prim = subset(dat_main, Cut.off.source == "Official")
res <- metafor::rma.mv(logOR ~ 1, V = seOR^2, random = ~ 1 | study_ID/es_id, data = dat_prim)
res
## 
## Multivariate Meta-Analysis Model (k = 41; method: REML)
## 
## Variance Components:
## 
##             estim    sqrt  nlvls  fixed          factor 
## sigma^2.1  0.0051  0.0717     39     no        study_ID 
## sigma^2.2  0.0000  0.0000     41     no  study_ID/es_id 
## 
## Test for Heterogeneity:
## Q(df = 40) = 70.0963, p-val = 0.0023
## 
## Model Results:
## 
## estimate      se    zval    pval    ci.lb   ci.ub 
##   0.0277  0.0194  1.4249  0.1542  -0.0104  0.0658    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Forest plot

Equivalence tests

tost = metafor::rma.mv(logOR ~ 1, V = seOR^2, random = ~ 1 | study_ID/es_id, data = dat_prim, level = 90)
## [1] "Because our analyses did not reveal a statistically significant effect of relative age on the persistence of ADHD, we performed equivalence tests to explore the smallest effect size that the data allow to reject. Equivalence tests show that even if relative age leads to a decrease in the persistence of ADHD, this effect is extremely small OR=0.996"

3. Moderation

Main model

 # {.tabset}
res_mod_i <- metafor::rma.mv(logOR ~ Cut_off_date, V = seOR^2, random = ~ 1 | study_ID/es_id, data = dat_main)
res_mod_i
## 
## Multivariate Meta-Analysis Model (k = 54; method: REML)
## 
## Variance Components:
## 
##             estim    sqrt  nlvls  fixed          factor 
## sigma^2.1  0.0035  0.0592     51     no        study_ID 
## sigma^2.2  0.0000  0.0000     54     no  study_ID/es_id 
## 
## Test for Residual Heterogeneity:
## QE(df = 52) = 77.1822, p-val = 0.0133
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.6922, p-val = 0.4054
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                 0.0273  0.0177   1.5433  0.1228  -0.0074  0.0620    
## Cut_off_dateProbable   -0.0301  0.0362  -0.8320  0.4054  -0.1012  0.0409    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
res_mod <- metafor::rma.mv(logOR ~ Cut_off_date - 1, V = seOR^2, random = ~ 1 | study_ID/es_id, data = dat_main)
res_mod
## 
## Multivariate Meta-Analysis Model (k = 54; method: REML)
## 
## Variance Components:
## 
##             estim    sqrt  nlvls  fixed          factor 
## sigma^2.1  0.0035  0.0592     51     no        study_ID 
## sigma^2.2  0.0000  0.0000     54     no  study_ID/es_id 
## 
## Test for Residual Heterogeneity:
## QE(df = 52) = 77.1822, p-val = 0.0133
## 
## Test of Moderators (coefficients 1:2):
## QM(df = 2) = 2.3897, p-val = 0.3027
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## Cut_off_dateOfficial    0.0273  0.0177   1.5433  0.1228  -0.0074  0.0620    
## Cut_off_dateProbable   -0.0028  0.0316  -0.0889  0.9292  -0.0648  0.0591    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Forest plot